[pull] main from microsoft:main

.semversioner/next-release/patch-20260219013931620740.json

@@ -0,0 +1,4 @@
+{
+  "type": "patch",
+  "description": "Add filtering, timestamp explosion, insert/count/remove/update operations to vector store API. Add top-level vector_size config to VectorStoreConfig."
+}

.semversioner/next-release/patch-20260220214632816094.json

@@ -0,0 +1,4 @@
+{
+  "type": "patch",
+  "description": "generate_text_embeddings streaming"
+}

dictionary.txt

@@ -27,6 +27,8 @@ dtypes
 ints
 genid
 isinstance
+ismatch
+ftype
 
 # Azure
 abfs

docs/examples_notebooks/index_migration_to_v1.ipynb

@@ -205,44 +205,23 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from graphrag.cache.factory import CacheFactory\n",
     "from graphrag.callbacks.noop_workflow_callbacks import NoopWorkflowCallbacks\n",
     "from graphrag.index.workflows.generate_text_embeddings import generate_text_embeddings\n",
-    "from graphrag.language_model.manager import ModelManager\n",
-    "from graphrag.tokenizer.get_tokenizer import get_tokenizer\n",
+    "from graphrag_cache import create_cache\n",
     "\n",
-    "# We only need to re-run the embeddings workflow, to ensure that embeddings for all required search fields are in place\n",
-    "# We'll construct the context and run this function flow directly to avoid everything else\n",
+    "# We only need to re-run the embeddings workflow, to ensure that embeddings\n",
+    "# for all required search fields are in place.\n",
+    "# We pass in the table_provider created earlier so that generate_text_embeddings\n",
+    "# reads the migrated tables we just wrote.\n",
     "\n",
-    "model_config = config.get_language_model_config(config.embed_text.model_id)\n",
     "callbacks = NoopWorkflowCallbacks()\n",
-    "cache_config = config.cache.model_dump()  # type: ignore\n",
-    "cache = CacheFactory().create_cache(\n",
-    "    cache_type=cache_config[\"type\"],  # type: ignore\n",
-    "    **cache_config,\n",
-    ")\n",
-    "model = ModelManager().get_or_create_embedding_model(\n",
-    "    name=\"text_embedding\",\n",
-    "    model_type=model_config.type,\n",
-    "    config=model_config,\n",
-    "    callbacks=callbacks,\n",
-    "    cache=cache,\n",
-    ")\n",
-    "\n",
-    "tokenizer = get_tokenizer(model_config)\n",
+    "cache = create_cache(config.cache)\n",
     "\n",
     "await generate_text_embeddings(\n",
-    "    text_units=final_text_units,\n",
-    "    entities=final_entities,\n",
-    "    community_reports=final_community_reports,\n",
+    "    config=config,\n",
+    "    table_provider=table_provider,\n",
+    "    cache=cache,\n",
     "    callbacks=callbacks,\n",
-    "    model=model,\n",
-    "    tokenizer=tokenizer,\n",
-    "    batch_size=config.embed_text.batch_size,\n",
-    "    batch_max_tokens=config.embed_text.batch_max_tokens,\n",
-    "    num_threads=model_config.concurrent_requests,\n",
-    "    vector_store_config=config.vector_store,\n",
-    "    embedded_fields=config.embed_text.names,\n",
     ")"
    ]
   }